In a data center in which multiple compute resources and storage resources communicate to execute workloads (e.g., applications, processes, threads, etc.), the speed at which the workloads can be processed may be limited by the speed at which data can be communicated among the resources. By compressing the data, such as with entropy encoding, the amount of data to be transferred can be reduced, thereby reducing the data transmission time. However, additional time is then spent to decompress the received compressed data.